Infrastructure notes by Jeevan Bobba on the decisions behind reliable systems: storage, access, recovery, and day-to-day operations.
Storage: ask about the failure first
When reviewing storage, I start with the workload and the recovery goal. Capacity and performance matter, but so do the time needed to rebuild a failed device and the steps needed to restore an accidentally deleted file.
RAID combines drives using techniques such as mirroring, striping, and parity. Each arrangement makes a different tradeoff between usable capacity, performance, and tolerance of drive failures. The right choice depends on the controller, drive characteristics, workload, and operational requirements.
RAID is not a substitute for a backup. Redundant storage can still be affected by deletion, corruption, ransomware, or the loss of the system that contains it. Keep recoverable copies and verify the restore process.
Practical guides from the archive
- Inspect SMB share permissions
- Find duplicate user principal names
- Report recently active Windows servers
- Review an archived Group Policy backup example
Make a change reviewable
Record the current configuration, describe the intended result, and choose checks that show whether the change worked. Keep the scope small enough to understand and identify a recovery path before changing production systems.
Browse general infrastructure articles or read about the author.